Distance metric learning for kernel density-based acoustic model under limited training data conditions.
Van Hai DoXiong XiaoEng Siong ChngHaizhou LiPublished in: APSIPA (2015)
Keyphrases
- distance metric learning
- training data
- metric learning
- distance metric
- component analysis
- kernel learning
- kernel matrix
- semi supervised
- image classification
- gender classification
- semidefinite programming
- training set
- kernel function
- decision trees
- feature space
- supervised learning
- kernel methods
- training process
- labeled data
- data sets
- training examples
- clustering algorithm
- semi supervised learning
- multiple kernel learning
- reproducing kernel hilbert space
- nonlinear dimensionality reduction
- nearest neighbor classification
- dimensionality reduction
- pairwise
- learning algorithm
- class labels
- preprocessing step
- training samples
- distance measure
- multi class
- support vector machine
- data points
- support vector
- feature extraction